2.6.39-ck2, bfs-0.406

These are patches designed to improve system responsiveness and interactivity with specific emphasis on the desktop, but suitable to any commodity hardware workload.


Apply to 2.6.39(.x):
http://www.kernel.org/pub/linux/kernel/people/ck/patches/2.6/2.6.39/2.6.39-ck2/patch-2.6.39-ck2.bz2

Ubuntu packages (2.6.39-ck1-3 is equivalent to 2.6.39-ck2):
http://ck.kolivas.org/patches/Ubuntu%20Packages/

Broken out tarball:
http://www.kernel.org/pub/linux/kernel/people/ck/patches/2.6/2.6.39/2.6.39-ck2/2.6.39-ck2-broken-out.tar.bz2

Discrete patches:
http://www.kernel.org/pub/linux/kernel/people/ck/patches/2.6/2.6.39/2.6.39-ck2/patches/

All -ck patches:
http://www.kernel.org/pub/linux/kernel/people/ck/patches/

BFS by itself:
http://ck.kolivas.org/patches/bfs/

Web:
http://kernel.kolivas.org

Code blog when I feel like it:
http://ck-hack.blogspot.com/

Each discrete patch contains a brief description of what it does at the top of the patch itself.


The only change from 2.6.39-ck1 is an upgrade to BFS CPU scheduler version 0.406. A bug that would cause hangs due to an incompatibility with the new block plug flushing code and BFS was fixed. For those who tried the "bfs404-test9" patch, this is only trivially different apart from the bfs version change.

Full patchlist:
2.6.39-sched-bfs-406.patch
sched-add-above-background-load-function.patch
mm-zero_swappiness.patch
mm-enable_swaptoken_only_when_swap_full.patch
mm-drop_swap_cache_aggressively.patch
mm-kswapd_inherit_prio-1.patch
mm-background_scan.patch
mm-idleprio_prio-1.patch
mm-lru_cache_add_lru_tail-1.patch
mm-decrease_default_dirty_ratio.patch
kconfig-expose_vmsplit_option.patch
hz-default_1000.patch
hz-no_default_250.patch
hz-raise_max.patch
preempt-desktop-tune.patch
ck2-version.patch

Please enjoy!
お楽しみください
--
-ck

lrzip tarball of all 40 linux-2.6 kernels

With the 2.6 linux kernel now officially finished, I'm providing a tarball of all 40 of the 3 point kernel releases as an lrzip tarball. This is a convenient way of getting all the releases in a relatively low bandwidth form, and previous archives of this nature have had a few downloads so I figured I'd complete the archive for those who want to download it and use it as an ad for lrzip:

163.9MB:
linux-2.6.0-2.6.39.tar.lrz

10.3GB:
linux-2.6.0-2.6.39.tar

Original file is a tarball of all 3 point linux kernel release tarballs 2.6.0 to 2.6.39 coming to a grand total of 10.3GB

Of course this is a plug for lrzip on something it is particularly good at compressing.

lrzip can be obtained here:
http://lrzip.kolivas.org

About this archive: It was compressed with lrzip version 0.606 on a quad core 3GHz core 2 on a relatively slow external USB2 hard drive with the following options:

lrzip -UL 9 linux-2.6.0-2.6.39.tar
Total time: 00:56:19.88

It would have compressed a lot faster without the -L 9 option, but given this is the "final" archive of 2.6, I figured I'd push it a bit further. Lrzip can compress it even further with zpaq as an option, but it makes decompression much slower so I'd personally find the archive less useful.

Of course someone will ask how it compares to xz, so for completion:

xz -9 linux-2.6.0-2.6.39.tar
Total time: 2:05:32.218

11067473920 linux-2.6.0-2.6.39.tar
1535618848 linux-2.6.0-2.6.39.tar.xz 13.8%
171879382 linux-2.6.0-2.6.39.tar.lrz 1.6%

EDIT:
I figured I'd do the rest as well.

Here is a tarball of all 161 stable 3 point releases from linux-1.0 to linux-2.6.39:

211MB:
linux-1.0-2.6.39.tar.lrz

19617064960 linux-1.0-2.6.39.tar
221368298 linux-1.0-2.6.39.tar.lrz 1.1%

As compressing this workload is mostly I/O bound I performed it on an SSD
drive this time:

linux-1.0-2.6.39.tar - Compression Ratio: 88.617. Average Compression Speed: 8.026MB/s.
Total time: 00:38:50.76

2.6.39 BFS test 9 - is this the one?

Hopefully this test patch should fix all the problems with BFS 404 on 2.6.39:

bfs404-test9.patch
Ubuntu Packages : grab the 2.6.39-ck1-3 package

Please report back if you haven't already! Thanks to everyone who has tested so far! Your feedback has been absolutely essential on this weird and wonderful bug.

2.6.39 BFS progress

TL;DR: 2.6.39 BFS fixed maybe?

After walking away from the code for a while, annoyed at the bug I couldn't track down, I had another good look at what might be happening. It appears that while the grq lock is dropped in schedule() to perform the block plug flush, a call to the task via try_to_wake_up may be missed entirely, leaving the task deactivated when it should actually keep running. Anyway, first tests from the people on these blog comments are reassuring.

Here is a cleaned up and slightly modified version of the "test8" patch that has so far been stable and shows to have fixed the problem for a handful of people:

Apply to 2.6.39-ck1 or 2.6.39 with BFS 404:
bfs404-recheck_unplugged.patch

In response to requests for packaged versions, I've uploaded a 2.6.39-ck1-2 ubuntu package which includes this change:
Ubuntu Packages

Please test and report back! If this fixes the problem, I'll be releasing it as ck2.